[00:00:15] Ghazenfer Mansoor: Hello and welcome to Lessons from the Leap. I’m your host, Ghazenfer Mansoor. On this show, I get to sit down with entrepreneurs, founders, and business leaders who talk about bold decisions, pivotal moments, and innovative ideas that shape their journeys.

This episode is brought to you by Technology Rivers. At Technology Rivers, we bring innovation through technology and AI to solve real world industry problems. If you wanna learn more about us and talk about any of your projects, head over to technologyrivers.com. Today on Lesson from the Leap, we have Sarah Cornett.

Sarah, welcome to the show. I’ll let Sarah introduce herself, share your background, and anything you want our listeners to know about you.

[00:00:59] Sarah Cornett: Oh, well thank you so much. I’m really excited to be here today and share my story. So I’m Sarah Cornett. I’m the founder and CEO of Global AI Advisors. We support clients on a global scale really supporting with advisory services and education around AI.

We support clients from industries like banking industry and finance to healthcare, to government industries. We support with things like developing AI strategies, finding the best opportunities to leverage AI. Establishing solid and safe AI governance programs. We also do quite a bit in the education space, so developing teams to leverage generative AI solutions, and also conducting AI leadership development training programs as well.

But prior to establishing global AI advisors I actually have experience working within Fortune 500 companies and implementing AI solutions. I’ve done this at leading Fortune 500 digital banks, and I’ve also done this with multinational banks and top airline clients as well. So taking in all of my experience in implementing AI solutions, I really wanted to start this business, start advising clients and companies from various industries and backgrounds and really setting businesses up for success when it comes to AI adoption and transformation. So that’s a little bit about me and my experience.

[00:02:24] Ghazenfer Mansoor: Cool. Thank you very much. So you worked across banking tech, and now we’re on global AI advisors. So what was the defining leap that pushed you to start your own AI advisory firm?

[00:02:40] Sarah Cornett: Absolutely. So I will say implementing AI at the scale and the, you know, at the enterprise level that I’ve worked in before at these really large technical companies was incredibly challenging and there were a number of, you know, challenges not only even on the data and technology side, but also with the change management aspect of AI implementation it’s incredibly challenging to implement these sophisticated, you know, solutions at such a high level, at the enterprise level, especially these highly regulated industries like banking and healthcare. So I just really wanted to take in all this information. I started seeing the kind of typical patterns and things that would come up with implementation and I, that really just sort of drove me to coming up with a strategic AI adoption methodology to really be able to support a number of clients at scale and also being able to do work internationally as well. 

So that’s how I’ve, you know, sort of taken all my expertise and everything that I’ve seen being a part of those large implementation projects and really being able to help a wider range of clients.

[00:03:49] Ghazenfer Mansoor: Thank you. So what is. The biggest gap you see between AI interest and actual AI execution inside the organization.

[00:04:01] Sarah Cornett: I would say there’s a few, but definitely starting with a solid AI strategy is one of the biggest gaps that I’m seeing with implementation, really seeing success with your AI solutions and implementation.

I, there was a recent MIT study that came out that, you know, did some research and, you know, really dug into these different implementations and pilot programs going on, especially around generative AI solutions and they found that around like 80 to 90% of AI projects actually failed to deliver expected business value, which is a shocking amount but it’s actually something that I’ve seen happen many times and organizations are sort of in this race and in this rush to become innovative and then get on the AI bandwagon, so to speak they’re sort of grasping at different solutions and trying things out and piloting different products, but they’re not coming up with solid, you know, business problems or business strategy behind the technology.

So I think that’s a huge gap, what we’re seeing in organizations today. But if you start with understanding your business objectives and understanding the strategy and the why behind adopting AI, you’re really going to be set up in a much more successful path forward when it comes to AI adoption.

[00:05:20] Ghazenfer Mansoor: Cool. You raise a good point about 80 to 90% of the AI projects failing, and in our software development business we do see the same thing because we are primarily helping healthcare customers, implementing AI, so we do see the same thing any thoughts, any on why those projects are failing and what could be done to stop that. I know you already said one thing, strategy for sure and that’s the mother of everything but is there any other thing you can share?

[00:06:02] Sarah Cornett: Definitely, I would say, you know, and I’m sure you’re aware of this as well, you know, it’s complicated. It can be complicated to, you know, adopt and onboard these solutions and from a technical perspective, right? You’ve gotta have your data in the right place, you’ve gotta have your technology systems talking together. But by and large, one of the largest, you know, challenges and hurdles when it comes to the successful adoption of ai. It’s actually on the people’s side of things.

So, you know, the technology and data is obviously going to be, you know, complicated to get that right. But the people that are really driving these projects and are behind, you know, some of these projects that’s, there’s a really large cultural shift in change management that needs to occur when you’re adopting these large AI solutions.

So, you know, traditionally technology projects, you know, they’re part of the technology team and they’re sort of pushed up from the tech side of things and you partner with your various business units and things like that to move forward with the projects. But you really need that solid buy-in and strategy from the business side and really having the business input in with these solutions and having the business really be bought in and really aligned with the AI solutions and really wanting these AI solutions as well. So I think on the people side, you know, there’s the change management aspect. 

There’s a big education aspect as well of, you know, having these businesses owners and business leaders understanding the why behind AI and getting behind these AI, you know, solutions because these business owners, you know, leaders are making decisions about which AI solutions we wanna bring on and which ones we wanna implement. But if they don’t have a solid education to understanding of the technology and what it can actually do and how it can actually add value to the organization, you’re definitely going to run into some challenges there.

[00:07:50] Ghazenfer Mansoor: A good point about change management. People are reluctant and this is a very common misconception as well because AI empowers people rather than replace and there’s a fear and in fact, it’s in many types of projects where usually the team is reluctant because they feel like if this thing comes, there will be the place.

But reality is different and if it is replaced, it’s gonna be replaced anyway. So it’s just a matter of delaying it. So, any thoughts on that? How to get the buy-in from your people or how do you, what strategies you can do to excite those people so that they’re willing to implement those AI projects rather than pushing back?

[00:08:45] Sarah Cornett: No, definitely yes and I think it all again, comes back to having a solid strategy and vision for AI for your business now. So thinking of, you know, whatever your business is, whatever industry you’re working in, like how do you see the future of your business with these AI solutions? Are you seeing, you know, solving any key problems throughout your organization? Or could there even be new products or services that you start to offer within this AI future that we’re living in now? 

So I think having a solid strategy at the highest level of your organization and pushing that down throughout your teams is really critical in just having overall buy-in within your AI vision for your organization, but then bringing your teams along with you, right?

You know, finding those opportunities and pin points. Are there specific teams that are running into, you know, challenges with keeping up with demand or any areas where they could really benefit from these AI solutions and really talk to them about how AI can really help them through their day-to-day, you know, business and work and things like this.

So finding those opportunities and workflows that are really going to matter and move the needle in terms of business value for your organization and so that way you can really work with those teams intimately, you know, helping them get training and education around these solutions. I think really doing a solid bit in upskilling across your organization is going to help bring those teams along and really get buy-in from those teams because, and again, you wanna be very clear on.

Are you cutting certain teams or replacing certain functions with AI solutions? Are you upskilling your teams now and having them work with the AI solutions? I think being very crystal clear about that is really going to help, you know, bridge this gap and help with the change management process.

Because like you said, you know, people are afraid of this technology. They’re afraid it’s going to replace them and their roles, but I think if you’re very clear about here is your role today of how that looks like with ai and we’re going to also help you get training and help you feel comfortable using these tools most effectively and safely as well. I think those are some good strategies to get buy-in from your teams.

[00:10:56] Ghazenfer Mansoor: Yeah, absolutely. Communication is key. Just like in any other part of the business that makes a huge difference how people foresee it. In our business, what we have seen. I think there’s another angle that we look at it, which we try with our customers, which is very successful. One of the things we suggest is start small and start slow.the moment you talk about over bringing ai, it creates a fear. 

But if you just do one small project, start showing the optimization, it could be starting with just creating those gpt, whether it’s your cloud, whether it’s your chatGPT or any, just start improving one process at a time. Once people start seeing it, they will get excited because you know, you are still giving them the same thing.

They already have 20 things on their plate. They are overwhelmed already. Yeah. They were supposed to work 40, 50, 60 hours. Now they could achieve the same work at much lower, and now they’re achieving a lot more. So suddenly they start to see their own productivity and they’re more excited because they’re delivering now more so doing one thing at a time or smaller pieces.

Suddenly you start seeing excitement and this is, I mean, we have seen in one of our client projects where we just delivered something in six week timeframe and then the whole team was excited and then they started coming into the meetings every week because now they saw something was working, which was not working before, and now they start bringing their problem.

And that was very interesting, like when you start seeing that team members are bringing, hey. What can we do with this problem? How can we do use technology and AI to solve this problem? And that was very interesting learning for us.

[00:13:04] Sarah Cornett: Yeah, no, that’s amazing and I couldn’t agree more. I think starting small is always really key with AI success.You know, I, it’s something I work with business leaders on this. When we’re developing strategies and prioritizing use cases, we first prioritize by expected business value generation, but then reprioritize based on feasibility. 

Because I’ve seen in these sort of big bang approaches where you’re trying to do too much, too fast. It can get really messy very quickly. So I think when you, like, as you said, just prioritize these small projects, get some comfortability and some confidence with embracing these solutions and really start to see the value like you mentioned there. 

I think when people are using the tools and they’re seeing how valuable they are and seeing, you know, how effective and efficient they can be, it just really starts to grow on and build on top of each other.So I think that’s an excellent approach and starting small. Getting those quick wins and starting to iterate on top of that.

[00:14:01] Ghazenfer Mansoor: Yeah. And you talked about in one of your substack you talked about five key strategies for competitive advantage. Can you put some light into that? And also, which single strategy delivers the highest return for companies right now?

[00:14:19] Sarah Cornett: Oh, I’ll have to think about that substack, I haven’t looked at that in a while exactly what those were competitive advantages with. I assess that was with AI Solutions and what you can do with them. But I mean, there’s a number of, you know, when I talk about business value and opportunities to leverage AI, you know, there’s most people think of the efficiency gains and, you know, really using it to cut time off of different projects you’re working on and different opportunities. A lot of people think of it as cost savings as well. So those opportunities to use it for cost savings, type of use cases, you know, those are ones that really quickly come to mind.

You think of, you know, some in the call center kind of use cases and things like this, but there are also incredible opportunities to drive revenue with AI technologies, which is something I think some businesses may not be as aware of, but there are incredible opportunities, like I think of personalization, kind of use cases and one-to-one communications with customers.

You can really get very specific in the communications and how you’re talking to customers and being very personalized, but you can use those in things like customer service types of communications, but also cross-sell upsell opportunities as well. So you can really start to drive value to your organization and you know, selling different products and services and solutions by doing some of those efforts there.

So when I’m thinking of competitive advantage in using these tools to really drive value. You know, those are some of the key places, but definitely don’t discount revenue generation because there are incredible opportunities to leverage AI in that space as well.

[00:15:58] Ghazenfer Mansoor: Absolutely in your AI advisory, you also talk about AI security and governance, right? Protecting investments. So can you talk about any governance missteps you have commonly seen in the enterprise AI programs?

[00:16:20] Sarah Cornett: Yes, definitely. I think first of all, if you’ve not stood up your AI governance program, you definitely need to start to take a look at this. You know, I think some organizations may get confused on, you know, the difference in data governance and AI governance, which many organizations have been going through some steps in, you know, quantifying, labeling, making sure their data is secure, safe, and reliable but what AI governance does is really taking in that data that’s fed into these systems and ensures your end-to-end AI systems are fully compliant. 

So if you’ve not taken steps in developing AI governance practices for your organizations, I highly encourage you to do this. Some organizations are further along with this, you know, different finance industries, financial companies are definitely developing in their AI governance.

Technology’s a bit further ahead as well, but some in healthcare, government education space are still trying to figure out what this looks like but the AI governance is very critical in securing your AI solutions and ensuring that they are reliable and trustworthy, but one of the first places I recommend you start is developing your team.

So developing a team within your organization from members of maybe your legal team, a compliance team member, your model validation team, data team, technology, cybersecurity. So really developing your core team that’s going to be responsible for AI governance with your organization and start first by understanding what regulations your industry made here to adhere to.

If you are a global organization, I do quite a bit of work in Europe with, and they have to comply with the EU AI Act. So understanding what regulations adhere to you based on your location, your country, where your customers are located. In the US it’s getting a little bit messy and a little bit challenging with some states are developing their own specific AI regulations.

I know New York and California have some things around deep fake technology specifically, but Colorado was starting to develop some really, you know, secure regulations around AI. So you really wanna start to map out, you know, there’s different policies and regulations are starting to build and start to starting to come up more and more every day.

So, good place to start is by understanding your specific industry, where your customers are located, and what are those regulations and policies and frameworks that you need to adhere to. So those are a really good place to start. But yeah, governance is incredibly key for adopting and implementing AI. 

[00:19:00] Ghazenfer Mansoor: Yeah. And do you also work on the regulated industry? I saw banking. I’m not sure about healthcare as well, like, have, do you see any difference in terms of how these organizations behave when it comes to AI?

[00:19:18] Sarah Cornett: Yeah, so I’ve actually been a part of establishing, I established an AI governance program within a Fortune 500 digital bank.

So I have experience in doing, this was several years back now, and most, I’d say most banks and financial institutions have some sort of governance, you know, program that they’ve established. I’ll say a lot of healthcare companies. I’m starting to talk to them now about developing some AI governance.

I think a lot of healthcare organizations I speak to, you know, they’re very you know, hesitant and wanna be really methodical about, you know, what makes the most sense to adopt AI solutions. You know, they’re very obviously critical of hipaa, right? Of, you know, how do we make sure, you know, we are compliant against our HIPAA regulations, what does that mean?

You know, they’re dealing with very sensitive information and data. So that’s where that’s what I hear from a lot of healthcare companies is they wanna be very mindful and very sensitive because they are dealing with incredibly sensitive information. So they wanna start with almost a governance approach first, and laying down that solid framework and guidelines of how do we make sure we are safely protecting our data and it’s being used in a safe way with these solutions first. So that’s what I’m seeing in healthcare. They’re starting to just get into this, but they’re being very methodical about how they go about this.

[00:20:47] Ghazenfer Mansoor: Thanks for sharing that.So our company, we work primarily with healthcare and that’s where a lot of those conversations about hipaa, PHI data and I think it’s the same in banking as well. So when you have personal data, and it’s also the same in legal and many other industries. So people are really concerned about their protected data and they don’t want LLMs to be trained on it.

So are there any strategies? you talk about when you, when it comes to whether it’s internal data, not necessarily be regulated, like even still, many companies have a proprietary data. They don’t want their people to be putting on the LMS for training and, but they still need it, that as part of their daily work, so they have to leverage AI for that.

What strategies do you recommend those companies or is that part of the scope of work you do or is that more goes toward technical companies implementation company like Technology Rivers?

[00:21:57] Sarah Cornett: Yeah, no, it’s definitely something I’ve supported clients through this as well and understanding, you know, that trace of data, right, of like how is that data being used and some of the enterprise, you know, license models.

There’s, you know, Chat GPT, like an enterprise license. I know a number of companies are rolling out Microsoft Copilot. I’ve done work as well with like IBM Watson. They have a very clear, you know, button to check to say, my data will not be used by these large language models and being used in any learning.

So some of those solutions are allow you to be like really locked down and only using your proprietary business information and it’s not being used to train any large language models. So there are some great, you know, tools, those enterprise solutions that come with that there are also great tools, what it’s called liminal. That helps you really put in your business policies around, you know, I can’t use any PII information. If I’m a bank. I can’t put in account numbers or any information like this if I’m a healthcare company. You know, these are the different data elements. I cannot include, you know, in any prompts or any information like this.

But what liminal will do will allow you to write the prompt, even if you make a mistake and accidentally write in an account number, for example, it will automatically mass that information within the prompt itself. So you can still continue on and continue with the prompt, but it’ll automatically mass that information.

So no sensitive data is being translated into those models whatsoever. So there are some really helpful tools in that to really help you know, embedding your either business policies or regulations, you know, whatever those specific guidelines are within the solutions that allow you to allow your end users to prop really safely. So those are some great solutions there.

[00:23:43] Ghazenfer Mansoor: No, that’s a really good insight. So I think the enterprise version, and I hear that a lot of companies are adapting Microsoft Co-pilot that’s becoming a bit more tendered for some of those organizations, not that others don’t like we work with healthcare, so we have BA science business Associate Agreement, which is required.

In any healthcare company if the data needs to be HIPAA compliant, so any organization which will have access to your data need to sign that to protect that there will be a zero retention policy and not being used for this audit, logging and all of those different rules. So and most of these LMS would sign Gemini, OpenAi. So we have worked with these and they have all signed it. So that gives some level of protection, but then still there’s data that you have your own data and that’s where we adding on when it comes to building those custom things, that’s where on these applications we have seen like approaches like RAG the retrieval augmented generation.

That makes a huge difference because now your own data is tokenized and put it in a vector database. You are searching that and only given that to LLM so that it searches the limited data and not getting access to your whole data. So things like that. So there are a lot of different approaches.

[00:25:14] Sarah Cornett: Definitely I’ve been hearing so much about these private sort of search functionalities. I think like Slack was in, like looking into, you know, in incorporating this technology, but similar to this RAG system you’re describing. But I like the  notion of, you know, especially these highly regulated industries, having your own internal sort of RAG systems.

You’re not relying on any third party tools so that you can vary safely and effectively search through your docents, search through your information, and not relying on these tools ’cause I think so many tools that I’m hearing about today are coming up with their own, you know, internal search functionality it seems like.

So, because it’s a huge need and it would be incredibly beneficial and very effective for business users, you know, trying to find the information that they need.

[00:25:58] Ghazenfer Mansoor: And how many of the cases, I mean, I don’t need an exact number, but have you seen adoption of private LLMs more than these public LLMs?  Or do you see, is it becoming more norm or less?

[00:26:15] Sarah Cornett: I’ve been hearing in some of these larger financial institutions, them building their own LLMs, and I’m like, okay, like, you know, I, and fully understand the reasoning behind this. You know, you’ve got all this sensitive information they wanna have, you know, full security of those systems end to end.

But I think that’s going to take a long time for these big banks and institutions to develop their own LLMs. I mean, we’re talking probably a minimum of 10 years to have these things fully realized. So, but I think that those industries as well, and those companies are also, you know, playing with the co-pilots and chatGPTs to sort of test these things out in a very secure way but knowing on the backend, they’re building in, you know, their own sort of in-house LLM tools to be used in the future because, you know, there’s a lot of efficiencies to gain today within these tools that are ready to go now. But if they’re thinking longer term of how do we secure this process fully end to end.

So that’s why they’re investing in some of these LLMs, but it’s, yeah. But it’s good to leverage the tools we have now, but just do that in the most safe and effective way.

[00:27:24] Ghazenfer Mansoor: Yeah. Now that we are talking on LLMs, so there are so many public LLMs, so ChatGPT, Gemini, Copilot, Perplexity, Grok, there are so many options. What are your opinion on those? How do you suggest your customers which one to use? Because definitely the default always go to claude or  ChatGPT so can you share some examples, give some ideas on where I should be using what?

[00:28:07] Sarah Cornett: Sure. Yes and you know, you listed a few tools there that are probably, you know, the most popular solutions. Everyone kind of knows about some of these solutions in the AI space, but there are in fact thousands of generative AI solutions in the market today. It’s really an astounding amount of solutions out there.

But what I always encourage you know, end users, you know, people who are looking to experiment and try some of these tools is think about the problem you’re trying to solve first. Like, how are you going to use these tools first and then try to find the best tool that works for you. I’m seeing, you know, the various family members and friends are playing with some of these tools, but they’re using them as almost a Google search functionality, which is not necessarily what you wanna use these tools for like I think of if I wanna search for something and I want, you know, really factual information that I know is from a trustworthy source, I’ll use a tool like Perplexity for this. 

If I wanna do really deep strategy work and maybe some personal work and getting different perspectives and opinions on something, I tend to use ChatGPT for things like this for helping with writing content, you know, around my newsletter or LinkedIn posts, I use Claude primarily for this. So different tools have different strengths for sure, and so it really depends on how you’re intending to use these tools and thinking about which tool actually makes the most sense to complete a specific task.

So I always encourage people to start with the problem you’re trying to solve first or the task you’re trying to complete, and then find the best solution.

[00:29:42] Ghazenfer Mansoor: Make sense, Yeah. So along that, obviously there are different models within those LLMs as well, so, and most people are still not familiar with it and I think as we were talking earlier, you mentioned that most people are still using it like a Google search. So yeah, if you wanna add anything to that would be great.

[00:30:09] Sarah Cornett: Yeah, definitely you know, they’re, these companies are, you know, or almost in a race with each other to build the most powerful LLM possible. So they’re always churning out these new LLMs and it’s very hard to keep up with you know, which is the most powerful, which one I should use for this, I always, as a rule of thumb, try to use.

The most powerful model, depending on, you know,which tool I’m using. Just go ahead and get to the highest one that you can because it’s typically gonna come with richer data, more information. You know, these models are just getting better and better but you wanna keep an eye out, you know, they are testing for things like AGI, and they’re starting to build to this sort of all general intelligence world. So some of them are trying to reason for themselves and make decisions for themselves. So some of those you have to be a little bit careful with how you’re interacting with them and just like understanding how to use those models specifically.

But I always just recommend just go with the highest model, you know, the best model that they have because they are getting better and better every day. When this technology was first released and brought onto the market, you know, they are riddled with issues like hallucinations. Those hallucination rates are way down across the board across these solutions now.

So that’s providing actually false information when you’re trying to ask a question. It’s not providing correct and accurate information. So but again, the models are getting better and better. So I encourage you to just try to use the most advanced model possible.

[00:31:37] Ghazenfer Mansoor: So you talked about hallucination. Is there anything users can do to reduce that hallucination? Obviously models are improving.

[00:31:46] Sarah Cornett: Yeah, it’s, I think it all comes back to the prompt. So being very specific in what question you’re asking and what you’re prompting and being really effective with your prompt. So you wanna try to think of, you know, how am I providing as much context as possible within this prompt, you know, am I asking the right question with this prompt?because really, you know, generative AI, at the end of the day, it comes from users prompting the tool and the tool giving a response back. So, you know, there’s ways you can really improve upon your prompts. You know, there’s great classes and prompt engineering to help you write the most effective prompts possible.

But I think that’s how the user’s going to get the best result back is to really come up with, you know, some sophisticated, that the best prompts possible to get the best result possible. So, you know, it can help some with the hallucinations as well if you’re asking the right questions but again a lot of that comes on the backend data of what some of these models were trained on. If they weren’t trained on the most up-to-date model as of November 2025, then that you’re not gonna get probably the most accurate results. So but yeah, I think from a user perspective, just try to come up with the best prompt possible when communicating with  AI.

[00:33:02] Ghazenfer Mansoor: So the tip is to be better in prompting, get some training on prompt engineering or learn even from ChatGPT.

[00:33:14] Sarah Cornett: Yeah. I think foundations and generative ai, you know, just general education around training is really critical just to really help you understand, you know, what these models mean, what are the different solutions out there and how I can use them and how I can be most effective when communicating and working with these tools.

[00:33:34] Ghazenfer Mansoor: Yep. So we’re talking about learning. I know AI is moving so fast that even the books that are coming or have come probably outdated. What is the best place for people to learn about AI? Like we talked about, for example, whether it’s prompt engineering, whether it’s about different models, naturally people will just go to the LLMs about learning.

About the other ones, which could be biased as well because ChatGPT is not gonna tell you too much about the other one. I mean, yes, it would, but maybe more biasness. So, any recommendations where people should go if they want to learn new techno, new ai? What is the best way for anybody to dig deeper into it?Knowing more than just the basics.

[00:34:30] Sarah Cornett: Definitely yes. I think, you know, education is so key in this new world of AI that we’re in so even if you are, you know, maybe you’re a student in college and you’re trying to figure out your major or you are a seasoned leader in an organization, you know, there’s something definitely to learn more about with this AI, with these different AI solutions.

So, you know, there are a number of great courses talked about prompting. You know, there are a number of great online courses for prompt engineering. So, you know, there’s incredible resources out there. You know, MIT Stanford has some great courses, edX has some incredible courses as well around prompting.

So helping you, you know, with the foundations in generative AI, just helping you understand, you know, the basics and understanding these generative AI solutions and how to most effectively prompt and use these tools. So the online courses are definitely something I would recommend. There are also some great courses and programs around leadership in AI because, you know, I think a lot of these, you know, leaders and organizations, you know, they’re trying to, you know, understand the best ways their teams can use AI.

And leaders are making decisions about which tools we invest in and infrastructure and data centers. You know, those are really incredibly complicated decisions that leaders are making and if these leaders don’t know the difference in generative AI and machine learning that’s gonna be a problem.

So, there’s definitely some leadership courses coming out as well. Courses around strategy and things like this cohort based courses. So I encourage you to look into those if you’re in more of a leadership role, but even, you know, just keeping up, like you mentioned, you know, there’s all these models coming out.

You know, there’s all these news about, you know, open AI enters a relationship with Nvidia and, you know, all these things are changing all the time. I personally leverage a lot of newsletters to keep up with all this technology and information that’s coming out. So there’s some really great newsletters you can subscribe to, you know, following on social media and things like this. So just keep up to date on all of the news and changes of when these models are released and things like this. So I encourage you to subscribe to some newsletters. I can come up with a few to send you and share with you some of those.

But Superhuman is one that definitely comes to mind as one that I am subscribed to. There’s always great information that comes out of that newsletter.

[00:36:50] Ghazenfer Mansoor: Yeah, thank you. If you share that, we’ll put the link in our podcast details. So what misconception about AI do you wish every business leader would stop believing?

[00:37:05] Sarah Cornett: Oh, that’s a really good question. A misconception about AI that every business leader would stop believing in? That’s a great question, I think one on one hand, I think probably the biggest thing is you can’t just plug it into the wall and it magically works and solves all your problems. I would say, I think a lot of people think if it’s something, you know, you buy it, you get it outta the box, you plug it into the wall, and it magically solves all of your problems.

Right you know, I’ve personally lived through many different implementations of AI solutions and it can be incredibly complicated to really roll these solutions out. You know, there’s, we touched on all these aspects throughout this conversation on, you know, from a strategy perspective to governance, to implementation to education and training to change management.

So there are so many complications that come with AI adoption implementation. So, you really wanna work with, you know, seasoned experts and people who have been there people who are really going to set your business up for success. When it comes to AI adoption and technology, I think a lot of businesses are trying to figure this out.

On their own they’re trying to, you know, go through this process internally, but you’re gonna just keep running into the same roadblocks and pin points that other organizations have done before. So, you know, I encourage leaders to look at different industry use cases studies, you know, working with different industry partners who have expertise in that specific industry, specific technology, and really just gaining learnings from each other and from these experts who have done this before and really listen to them and like, let them help you through this process because it’s not, you can’t just plug things into the wall and they magically work. There’s a lot of thought and a lot of planning and strategy and work that needs to be done to set your businesses up for success when it comes to AI transformation.

[00:38:58] Ghazenfer Mansoor: Well, one last question. What emerging AI capabilities excite you the most for the future of customer experience and decisioning?

[00:39:09] Sarah Cornett: Ooh, that’s a really good question. I think I’m most excited for multimodal AI actually. So, and multimodal ai, for those who are listening it is where you’re getting different data points, different incision points for information and making decisions based on an information.

So the way I like to describe it is imagine from a healthcare use case perspective, right? You know, some of us have wearable devices, right? You know, you’re getting data and information from that device. You’ve got information from your phone, maybe you have a smart scale at home, so you’re able to use all this different information about you and being able to make decisions about you personally from a healthcare perspective.

So I think that’s gonna be a really incredible opportunities that are gonna continue to evolve for the multimodal AI. I’m seeing that popping up a little bit more in 2026 is gonna be something really interesting and something to keep aware of because there’s incredible opportunities from a customer experience perspective, as you said, to use this kind of technology.

So I think multimodal AI is really gonna start to develop some really interesting, unique use cases and opportunities for customer experience.

[00:40:27] Ghazenfer Mansoor: Glad to know that I will check that out as well and we’ll share with the audience as well. So, Sarah, it’s been great having you on the lessons from the leap before we sign up, two quick things. One, can you share where people can find you and follow your work in AI innovation and enterprise transformation?

[00:40:46] Sarah Cornett: Absolutely, yes. So I am located in Charlotte, North Carolina, but I do business on a global scale. So definitely follow me on LinkedIn.

That’s where I’m probably the most active. I can share with you the links to my LinkedIn, my website has more information about my business and the work that we do. So, but if there’s anything at all. Happy to connect with you. You know, share some thoughts about different strategic perspectives, you know, different industry use cases and things like this.

But yeah, I’m happy to share those links within the meeting notes with you. But LinkedIn is definitely the best place to keep in touch.

[00:41:22] Ghazenfer Mansoor: Well, and is there any success story that you wanna share with our audience? Does not have to be specific. Could be anonymous.

[00:41:31] Sarah Cornett: Ooh. Okay. This is coming at a really good time. Actually. So I sit on an IT advisory board for Mecklenburg County, which is the surrounding county in the Charlotte, North Carolina area. We just did our annual report, which I was one of the presenters and speakers of this. So we presented to the board of county Commissioners of Mecklenburg County.

We just did this just this week. The presentation went extremely well, but what we do is provide guidance in the IT department for the county in some ways. We can, you know, find some real efficiencies and provide recommendations and I spoke on AI transformation and leading in AI transformation for Mecklenburg County, and it was received incredibly well.

The board had really thoughtful questions, but the board seems, you know, very eager to continue in this mission for AI transformation, being really thoughtful about this and so they were very happy and confident with our report, and we’re very grateful for us to helping lead the county into the next phase of AI transformation.So definitely was a highlight for me this week.

[00:42:41] Ghazenfer Mansoor: Awesome. Good week then. Thanks again. Thanks for having you on Lessons From the Leap, I’m sure our audience have got, learned so much more about AI, a lot more smarter after listening to this podcast. So thanks again for being on the show.

[00:43:01] Sarah Cornett: Oh, thank you for having me.